A power grid dynamic parameter identification system based on big data

Abstract

The present invention discloses a power grid dynamic parameter identification system based on big data, comprising a data acquisition module, a parameter detection module and a transmission management module, characterized in that: the data acquisition module is used to collect the operation status data of the power grid system, collect the historical data of the power grid system and various parameters in the power grid line, the parameter detection module is used to analyze the historical data of the power grid, train the parameter identification theoretical model, analyze the influencing factors affecting the operation status of the power grid system, and analyze whether the real-time operation status parameters of the power grid system are normal according to the influencing factors, the transmission management module is used to pre-process the collected data, and allocate fog nodes to the data according to the real-time bandwidth of the fog nodes and the remaining capacity of the data cache, the data acquisition module, the parameter detection module and the transmission management module are connected to each other in communication, and the present invention has the characteristics of improving the accuracy of the system and the efficiency of data transmission.

Description

Technical Field

[0001] The present invention relates to the technical field of parameter identification, and in particular to a power grid dynamic parameter identification system based on big data. Background Art

[0002] With the continuous increase of distributed power sources in the distribution network system, a large number of distributed power sources have been connected to the power grid, causing changes in the topology structure and load changes of the power grid. Most of the existing technologies use the method of setting threshold comparison to determine whether the power system has a fault, and realize the parameter identification of the power grid by tracking the dynamic changes of power grid parameters in real time and comparing the dynamic changes of parameters. However, due to the variability of the power system operating environment and the insignificant changes in the power system operating environment, it is easy to cause the changes in power grid parameters to be insignificant, resulting in misjudgment and omission of the system, affecting the accuracy of the system's identification of power grid dynamic parameters, resulting in poor accuracy of system power regulation. Moreover, due to the huge amount of data in the power grid system and limited system cache resources, slow data transmission is likely to occur during the transmission process, resulting in poor data real-time performance and low transmission efficiency. Therefore, it is very necessary to design a power grid dynamic parameter identification system based on big data to improve system accuracy and data transmission efficiency. Summary of the Invention

[0003] The purpose of the present invention is to provide a power grid dynamic parameter identification system based on big data to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a big data-based power grid dynamic parameter identification system, comprising a data acquisition module, a parameter detection module and a transmission management module, characterized in that: the data acquisition module is used to collect operating status data of the power grid system, collect historical data of the power grid system and various parameters in the power grid lines; the parameter detection module is used to analyze the historical power grid data, train a parameter identification theoretical model, analyze the influencing factors affecting the operating status of the power grid system, and analyze whether the real-time operating status parameters of the power grid system are normal based on the influencing factors; the transmission management module is used to pre-process the collected data and allocate fog nodes to the data based on the real-time bandwidth of the fog nodes and the remaining capacity of the data cache; the data acquisition module, parameter detection module and transmission management module are communicatively connected to each other;

[0005] The model training module includes a feature extraction submodule, a model training submodule and a state prediction submodule. The feature extraction submodule is used to analyze sample data and extract sample data features. The model training submodule is used to train a parameter identification theoretical model based on the extracted sample data features. The state prediction submodule is used to predict the operating state of the power grid based on the parameter identification theoretical model.

[0006] The parameter identification module includes an influencing factor analysis submodule and a time-frequency analysis submodule. The influencing factor analysis submodule is used to analyze the impact of the current topology, load and distributed power supply grid connection on energy consumption and grid dynamic parameters. The time-frequency analysis submodule is used to further identify dynamic parameters using time domain and frequency domain.

[0007] According to the above technical solution, the data acquisition module includes a data acquisition module, a historical data entry module and a sensor module. The data acquisition module is used to collect real-time operating status data of the power grid system, the historical data entry module is used to enter the historical status data of the power grid system into the system, and the sensor module is used to collect various parameters of the lines in the power grid system in real time.

[0008] According to the above technical solution, the parameter detection module includes a model training module and a parameter identification module. The model training module is used to analyze the historical state data of the power grid and train the parameter identification theoretical model according to the analysis results.

[0009] According to the above technical solution, the transmission management module includes a data preprocessing module and a port management module. The data preprocessing module is used to preprocess the collected data, and the port management module is used to manage the channels for data transmission in the fog node.

[0010] According to the above technical solution, the transmission management module further includes a node allocation module, which is used to analyze the real-time bandwidth and processing speed of the fog nodes and allocate fog nodes to the data according to the analysis results.

[0011] According to the above technical solution, the operation method of the power grid dynamic parameter identification system mainly includes the following steps:

[0012] Step S1: Periodically collect the operating status data of the power grid system through the data acquisition module, enter the historical data of the power grid system into the system through the historical data entry module, and collect various parameters of the power grid line in real time through the sensor module;

[0013] Step S2: When training the theoretical model, the system starts the model training module, starts analyzing the sample data, extracts the sample data features, and trains the parameters to identify the theoretical model based on the extracted sample data;

[0014] Step S3: After the model training is completed, the system sends an electrical signal to start the parameter identification module, starts analyzing the factors affecting the parameters of the power grid system, and identifies whether the dynamic parameters of the power grid are normal based on the factors;

[0015] Step S4: When data is transmitted, the system starts the transmission management module, begins to analyze the data volume in the power grid system, the bandwidth and processing speed of the fog nodes, and adjusts the fog nodes to which data is allocated based on the analysis results.

[0016] According to the above technical solution, step S2 further includes the following steps:

[0017] Step S21: Retrieving historical status data in the power grid system, integrating the historical status data in the power grid system, and preprocessing the data to remove redundant data in the merged data, identify the values ​​of various parameters in the historical data, and remove incomplete and repeated parameter items in the historical data;

[0018] Step S22: Obtain the pre-processed data, identify the timestamps in the data, anchor the data within the most recent period, arrange the remaining data according to the timestamps, and merge the parameters at the same time into a parameter set. Search the database according to the timestamps, retrieve the data parameter value range corresponding to the timestamps, and perform a secondary screening on the data. If there is a parameter value in the merged data that is greater than the maximum threshold or less than the minimum threshold, the data is discarded. Otherwise, the system continues to detect;

[0019] Step S23: Obtain the processed data, use the processed data as sample data, retrieve the corresponding power grid operation status in the database according to each parameter item, retrieve the theoretical model selected in the system, inject the processed data into the theoretical model for training, and obtain the parameter identification theoretical model. Substitute the data in the most recent anchored period into the parameter identification theoretical model to obtain the predicted power grid system parameter status.

[0020] According to the above technical solution, step S3 further includes the following steps:

[0021] Step S31: retrieve historical data from the power grid system, obtain the parameter set corresponding to the target timestamp, identify the parameter values ​​corresponding to the target timestamp, and calculate the discreteness of the parameter set corresponding to the target timestamp using the formula Where i = 1, 2, 3.......n, P represents the discreteness of the parameter set corresponding to the target timestamp, M represents the parameter value corresponding to the target timestamp, Represents the average value of the parameter set corresponding to the target timestamp, and compares it with the threshold set by the system. When the dispersion is less than the threshold, the database is searched according to the dispersion to retrieve the corresponding dispersion influence coefficient α on the parameter;

[0022] Step S32: When the discreteness is greater than the threshold, a coordinate system is established to identify the frequency of occurrence of each parameter value in the parameter set at the target timestamp, a line graph model is established based on the frequency of occurrence of each parameter value in the parameter set at the target timestamp, the change characteristics of the line graph model are extracted, and the change characteristics of the line graphs of the parameter sets of each target timestamp are fused.

[0023] According to the above technical solution, step S32 further includes the following steps:

[0024] Step S321: Retrieve the real-time operating status data of the device and compare it with the device operating status data predicted by the parameter identification theoretical model. If the real-time operating status of the device does not match the predicted operating status, identify the power grid system topology and grid-connected data. If the current power system topology and grid-connected power have not changed, mark the current device as faulty. Otherwise, continue detection.

[0025] Step S322: Retrieve the parameter prediction results of the parameter identification theoretical model and the results of the time-frequency analysis, calculate the difference between the parameter values ​​of the above prediction results, and if the difference between the parameters of the prediction results is less than the threshold, select the average value of the prediction results as the evaluation value, retrieve the real-time power grid parameters, calculate the difference with the evaluation value, and if the difference is less than the minimum threshold set by the system, mark the power system operation as normal, otherwise mark the power system operation as faulty, otherwise compare the actual parameters with the parameter prediction results of the parameter identification theoretical model and the results of the time-frequency analysis respectively;

[0026] Step S323: obtaining the number of devices and grid-connected power in the real-time power system topology, and retrieving the corresponding influencing coefficients β and κ on the power grid parameters in the database according to the number of devices and grid-connected power in the power system topology;

[0027] Step S324: When the difference between the parameter value of the parameter prediction result of the parameter identification theoretical model and the parameter value of the result of the time-frequency analysis is greater than the threshold, the result of the time-frequency analysis is obtained, and the actual parameter value corresponding to the analysis result is calculated by the time-frequency analysis formula Q=T·α·β·κ, where Q represents the actual parameter value corresponding to the time-frequency analysis result, and the real-time power grid parameter value is obtained. The differences F1 and F2 between the real-time power grid parameter value and the actual parameter value and the parameter value of the parameter prediction result of the parameter identification theoretical model are calculated respectively. If F1 and F2 are both less than the threshold, the current power system state is marked as normal. If F1 is less than the threshold and F2 is greater than the threshold, the time domain analysis model is marked as faulty. If F1 is greater than the threshold and F2 is less than the threshold, the parameter identification theoretical model is iteratively updated. Otherwise, the power system is marked as faulty. By comparing with the prediction results of the parameter identification theoretical model and the time domain analysis results respectively, it is possible to further reduce contingency, avoid large deviations in the system identification results, and further improve the accuracy of the system.

[0028] According to the above technical solution, in step S4, the fog nodes around the data acquisition device are retrieved, the remaining bandwidth and data cache remaining capacity of the target fog node are identified, the corresponding upload speed V1 in the database is retrieved according to the bandwidth, and the corresponding cache speed influence coefficient μ in the database is retrieved according to the data cache remaining capacity. The data collected by the data acquisition device is obtained, the number of bytes of the data is identified, and the amount of fog node cache occupied by the data to be transmitted is calculated by the formula Where H represents the amount of data to be transmitted that occupies the fog node cache, W represents the number of bytes of data to be transmitted, and V2 represents the cache speed of the target fog node. If the remaining capacity of the target fog node's data cache is less than H, a fog node with a data cache remaining capacity that meets the conditions is retrieved from the neighboring fog nodes and marked as the first fog node. The fog node port code in the data port of the data acquisition device is cleared, and the port code in the first fog node is identified for connection. After the connection is completed, data transmission is performed.

[0029] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention can reduce the impact of abnormal data on the accuracy of the parameter identification theoretical model by eliminating abnormal data in the data, thereby improving the accuracy of the parameter identification model prediction; by analyzing the discreteness of the parameter set at the target timestamp and using different methods to predict the power grid parameters according to different discreteness, it can avoid the large deviation of the prediction results caused by the solidified program, thereby improving the accuracy of the prediction results; by analyzing whether the topology structure and the grid-connected power have changed, it can avoid the influence of the topology structure and the grid-connected power on the operating status of the equipment, thereby causing the system to misjudge, and further improve the accuracy of the system; by comparing the prediction results of the two methods, it can avoid the large deviation of the prediction results caused by the slow update of the theoretical model, thereby reducing the accuracy of the system, thereby greatly improving the accuracy of the system; by analyzing whether the target fog node can accommodate the data to be transmitted, it can avoid the fog node being unable to accommodate the amount of data during the transmission process, resulting in insufficient resources of the fog node and inability to transmit data, resulting in slow data transmission, or even data loss, thereby further improving the data transmission efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0031] Figure 1 It is a schematic diagram of the system module composition of the present invention;. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] See also Figure 1 The present invention provides a technical solution: a power grid dynamic parameter identification system based on big data, including a data acquisition module, a parameter detection module and a transmission management module, characterized in that: the data acquisition module is used to collect the operating status data of the power grid system, collect the historical data of the power grid system and various parameters in the power grid line; the parameter detection module is used to analyze the historical data of the power grid, train the parameter identification theoretical model, analyze the influencing factors affecting the operating status of the power grid system, and analyze whether the real-time operating status parameters of the power grid system are normal based on the influencing factors; the transmission management module is used to pre-process the collected data and allocate fog nodes to the data according to the real-time bandwidth of the fog nodes and the remaining capacity of the data cache; the data acquisition module, the parameter detection module and the transmission management module are communicatively connected to each other;

[0034] The model training module includes a feature extraction submodule, a model training submodule and a state prediction submodule. The feature extraction submodule is used to analyze sample data and extract sample data features. The model training submodule is used to train a parameter identification theoretical model based on the extracted sample data features. The state prediction submodule is used to predict the operating state of the power grid based on the parameter identification theoretical model.

[0035] The parameter identification module includes an influencing factor analysis submodule and a time-frequency analysis submodule. The influencing factor analysis submodule is used to analyze the impact of the current topology, load and distributed power supply grid connection on energy consumption and grid dynamic parameters. The time-frequency analysis submodule is used to further identify the dynamic parameters using the time domain and frequency domain.

[0036] The data acquisition module includes a data acquisition module, a historical data entry module and a sensor module. The data acquisition module is used to collect real-time operating status data of the power grid system. The historical data entry module is used to enter the historical status data of the power grid system into the system. The sensor module is used to collect various parameters of the lines in the power grid system in real time.

[0037] The parameter detection module includes a model training module and a parameter identification module. The model training module is used to analyze the historical status data of the power grid and train the parameter identification theoretical model based on the analysis results.

[0038] The transmission management module includes a data preprocessing module and a port management module. The data preprocessing module is used to preprocess the collected data, and the port management module is used to manage the channels for data transmission in the fog node.

[0039] The transmission management module also includes a node allocation module, which is used to analyze the real-time bandwidth and processing speed of the fog nodes and allocate fog nodes to the data based on the analysis results.

[0040] The operation method of the power grid dynamic parameter identification system mainly includes the following steps:

[0041] Step S1: Periodically collect the operating status data of the power grid system through the data acquisition module, enter the historical data of the power grid system into the system through the historical data entry module, and collect various parameters of the power grid line in real time through the sensor module;

[0042] Step S2: When training the theoretical model, the system starts the model training module, starts analyzing the sample data, extracts the sample data features, and trains the parameters to identify the theoretical model based on the extracted sample data;

[0043] Step S3: After the model training is completed, the system sends an electrical signal to start the parameter identification module, starts analyzing the factors affecting the parameters of the power grid system, and identifies whether the dynamic parameters of the power grid are normal based on the factors;

[0044] Step S4: When data is transmitted, the system starts the transmission management module, begins to analyze the data volume in the power grid system, the bandwidth and processing speed of the fog nodes, and adjusts the fog nodes to which data is allocated based on the analysis results.

[0045] Step S2 further includes the following steps:

[0046] Step S21: Retrieving historical status data in the power grid system, integrating the historical status data in the power grid system, and preprocessing the data to remove redundant data in the merged data, identify the values ​​of various parameters in the historical data, and remove incomplete and repeated parameter items in the historical data;

[0047] Step S22: Obtain the preprocessed data, identify the timestamps in the data, anchor the data within the most recent period, arrange the remaining data according to the timestamps, and merge the parameters at the same time into a parameter set. Search the database according to the timestamps, retrieve the data parameter value range corresponding to the timestamps, and perform a secondary screening on the data. If there is a parameter value greater than the maximum threshold or less than the minimum threshold in the merged data, the data will be eliminated. Otherwise, the system will continue to detect. By eliminating abnormal data in the data, the impact of abnormal data on the accuracy of the parameter identification theoretical model can be reduced, thereby improving the accuracy of the parameter identification model prediction;

[0048] Step S23: Obtain the processed data, use the processed data as sample data, retrieve the corresponding power grid operation status in the database according to each parameter item, retrieve the theoretical model selected in the system, inject the processed data into the theoretical model for training, and obtain the parameter identification theoretical model. Substitute the data in the most recent anchored period into the parameter identification theoretical model to obtain the predicted power grid system parameter status.

[0049] Step S3 further includes the following steps:

[0050] Step S31: retrieve historical data from the power grid system, obtain the parameter set corresponding to the target timestamp, identify the parameter values ​​corresponding to the target timestamp, and calculate the discreteness of the parameter set corresponding to the target timestamp using the formula Where i = 1, 2, 3.......n, P represents the discreteness of the parameter set corresponding to the target timestamp, M represents the parameter value corresponding to the target timestamp, Represents the average value of the parameter set corresponding to the target timestamp, and compares it with the threshold set by the system. When the dispersion is less than the threshold, the database is searched according to the dispersion to retrieve the corresponding dispersion influence coefficient α on the parameter;

[0051] Step S32: When the discreteness is greater than the threshold, a coordinate system is established to identify the frequency of occurrence of each parameter value in the parameter set at the target timestamp, a line graph model is established based on the frequency of occurrence of each parameter value in the parameter set at the target timestamp, and the change characteristics of the line graph model are extracted. The change characteristics of the line graphs of the parameter sets of each target timestamp are fused. By analyzing the discreteness of the parameter set at the target timestamp and using different methods to predict the power grid parameters according to different discreteness, it is possible to avoid large deviations in the prediction results caused by the solidified program, thereby improving the accuracy of the prediction results.

[0052] Step S32 further includes the following steps:

[0053] Step S321: Retrieve the real-time operating status data of the device and compare it with the device operating status data predicted by the parameter identification theoretical model. If the real-time operating status of the device does not match the predicted operating status, identify the power grid system topology and grid-connected data. If the current power system topology and grid-connected power have not changed, mark the current device as faulty. Otherwise, continue detection. By analyzing whether the power system topology and grid-connected power have changed, the influence of the topology and grid-connected power on the device operating status, which may cause the system to make a misjudgment, can be avoided, thereby further improving the accuracy of the system.

[0054] Step S322: retrieve the parameter prediction results of the parameter identification theoretical model and the results of the time-frequency analysis, calculate the difference between the parameter values ​​of the above prediction results, if the difference between the parameters of the prediction results is less than the threshold, select the average value of the prediction results as the judgment value, retrieve the real-time power grid parameters, calculate the difference with the judgment value, if the difference is less than the minimum threshold set by the system, mark the power system operation as normal, otherwise mark the power system operation as faulty, otherwise compare the actual parameters with the parameter prediction results of the parameter identification theoretical model and the results of the time-frequency analysis respectively. By comparing the prediction results of the two methods, it is possible to avoid the large deviation of the prediction results caused by the slow update of the theoretical model, resulting in a reduction in system accuracy, thereby greatly improving the accuracy of the system;

[0055] Step S323: obtaining the number of devices and grid-connected power in the real-time power system topology, and retrieving the corresponding influencing coefficients β and κ on the power grid parameters in the database according to the number of devices and grid-connected power in the power system topology;

[0056] Step S324: When the difference between the parameter value of the parameter prediction result of the parameter identification theoretical model and the parameter value of the result of the time-frequency analysis is greater than the threshold, the result of the time-frequency analysis is obtained, and the actual parameter value corresponding to the analysis result is calculated by the time-frequency analysis formula Q=T·α·β·κ, where Q represents the actual parameter value corresponding to the time-frequency analysis result, and the real-time power grid parameter value is obtained. The differences F1 and F2 between the real-time power grid parameter value and the actual parameter value and the parameter value of the parameter prediction result of the parameter identification theoretical model are calculated respectively. If F1 and F2 are both less than the threshold, the current power system state is marked as normal. If F1 is less than the threshold and F2 is greater than the threshold, the time domain analysis model is marked as faulty. If F1 is greater than the threshold and F2 is less than the threshold, it means that the parameter identification theoretical model needs to be adjusted, and the parameter identification theoretical model is iteratively updated. Otherwise, the power system is marked as faulty. By comparing with the prediction results of the parameter identification theoretical model and the time domain analysis results respectively, it is possible to further reduce contingency, avoid large deviations in the system identification results, and further improve the accuracy of the system.

[0057] In step S4, the fog nodes around the data acquisition device are retrieved, the remaining bandwidth and data cache remaining capacity of the target fog node are identified, the corresponding upload speed V1 in the database is retrieved according to the bandwidth, and the corresponding cache speed influence coefficient μ in the database is retrieved according to the data cache remaining capacity. The data collected by the data acquisition device is obtained, the number of bytes of the data is identified, and the amount of fog node cache occupied by the data to be transmitted is calculated by the formula

[0058] Where H represents the amount of data to be transmitted that occupies the fog node cache, W represents the number of bytes of data to be transmitted, and V2 represents the cache speed of the target fog node. If the remaining data cache capacity of the target fog node is less than H, a fog node with a qualified remaining data cache capacity among the neighboring fog nodes is retrieved and marked as the first fog node. The fog node port code in the data port of the data acquisition device is cleared, and the port code in the first fog node is identified for connection. After the connection is completed, data transmission is performed. By analyzing whether the target fog node can accommodate the data to be transmitted, it is possible to avoid the situation where the fog node cannot accommodate the data volume during the transmission process, resulting in insufficient resources of the fog node and inability to transmit data, slow data transmission, or even data loss, further improving the data transmission efficiency of the system.

[0059] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0060] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A power grid dynamic parameter identification system based on big data, comprising a data acquisition module, a parameter detection module and a transmission management module, characterized in that: The data acquisition module is used to collect the operation status data of the power grid system, collect the historical data of the power grid system and various parameters in the power grid line. The parameter detection module is used to analyze the historical data of the power grid, train the parameter identification theoretical model, analyze the influencing factors affecting the operation status of the power grid system, and analyze whether the real-time operation status parameters of the power grid system are normal according to the influencing factors. The transmission management module is used to pre-process the collected data and allocate fog nodes to the data according to the real-time bandwidth and processing speed of the fog nodes. The data acquisition module, the parameter detection module and the transmission management module are connected to each other in communication; The model training module includes a feature extraction submodule, a model training submodule and a state prediction submodule. The feature extraction submodule is used to analyze sample data and extract sample data features. The model training submodule is used to train a parameter identification theoretical model according to the extracted sample data features. The state prediction submodule is used to predict the operating state of the power grid according to the parameter identification theoretical model. The parameter identification module includes an influencing factor analysis submodule and a time-frequency analysis submodule. The influencing factor analysis submodule is used to analyze the impact of the current topology, load and distributed power supply grid connection on energy consumption and grid dynamic parameters. The time-frequency analysis submodule is used to further identify dynamic parameters using time domain and frequency domain.

2. The big data-based power grid dynamic parameter identification system according to claim 1, characterized in that: The data acquisition module includes a data collection module, a historical data entry module and a sensor module. The data collection module is used to collect real-time operating status data of the power grid system, the historical data entry module is used to enter the historical status data of the power grid system into the system, and the sensor module is used to collect various parameters of the lines in the power grid system in real time.

3. The power grid dynamic parameter identification system based on big data according to claim 2 is characterized by: The parameter detection module includes a model training module and a parameter identification module. The model training module is used to analyze the historical state data of the power grid and train the parameter identification theoretical model according to the analysis results.

4. The big data-based power grid dynamic parameter identification system according to claim 3 is characterized by: The transmission management module includes a data preprocessing module and a port management module. The data preprocessing module is used to preprocess the collected data, and the port management module is used to manage the channel for data transmission in the fog node.

5. The big data-based power grid dynamic parameter identification system according to claim 4, characterized in that: The transmission management module also includes a node allocation module, which is used to analyze the real-time bandwidth and processing speed of the fog nodes and allocate fog nodes to the data according to the analysis results.

6. The big data-based power grid dynamic parameter identification system according to claim 5, characterized in that: The operation method of the power grid dynamic parameter identification system mainly includes the following steps: Step S1: periodically collect the operating status data in the power grid system through the data acquisition module, enter the historical data in the power grid system into the system through the historical data entry module, and collect various parameters in the power grid line in real time through the sensor module; Step S2: When training the theoretical model, the system starts the model training module, starts analyzing the sample data, extracts the sample data features, and trains the parameters to identify the theoretical model based on the extracted sample data; Step S3: After the model training is completed, the system sends an electrical signal to start the parameter identification module, starts to analyze the influencing factors that affect the parameters of the power grid system, and identifies whether the dynamic parameters of the power grid are normal based on the influencing factors; Step S4: When data is transmitted, the system starts the transmission management module, begins to analyze the amount of data in the power grid system, the bandwidth and processing speed of the fog nodes, and adjusts the fog nodes to which data is allocated according to the analysis results.

7. The big data-based power grid dynamic parameter identification system according to claim 6, characterized in that: The step S2 further comprises the following steps: Step S21: Retrieving historical status data in the power grid system, integrating the historical status data in the power grid system, and preprocessing the data to remove redundant data in the merged data, identify the values ​​of various parameters in the historical data, and remove incomplete and repeated parameter items in the historical data; Step S22: Obtain the preprocessed data, identify the timestamp in the data, anchor the data in the most recent period, arrange the remaining data according to the timestamp, and merge the parameters at the same time into a parameter set, search the database according to the timestamp, call the data parameter value range corresponding to the timestamp, and perform a secondary screening on the data. If there is a parameter value in the merged data that is greater than the maximum threshold or less than the minimum threshold, the data will be removed, otherwise the system will continue to detect; Step S23: Obtain the processed data, use the processed data as sample data, retrieve the corresponding power grid operation status in the database according to each parameter item, retrieve the theoretical model selected in the system, inject the processed data into the theoretical model for training, and obtain the parameter identification theoretical model. Substitute the data in the most recent anchored period into the parameter identification theoretical model to obtain the predicted power grid system parameter status.

8. The big data-based power grid dynamic parameter identification system according to claim 7, characterized in that: The step S3 further comprises the following steps: Step S31: retrieve historical data in the power grid system, obtain the parameter set corresponding to the target timestamp, identify the parameter values ​​corresponding to the target timestamp, and calculate the discreteness of the parameter set corresponding to the target timestamp through the formula Where i = 1, 2, 3.......n, P represents the discreteness of the parameter set corresponding to the target timestamp, M represents the parameter value corresponding to the target timestamp, Indicates the average value of the parameter set corresponding to the target timestamp, and compares it with the threshold set by the system. When the discreteness is less than the threshold, the database is searched according to the discreteness to retrieve the corresponding discreteness influence coefficient α on the parameter; Step S32: When the discreteness is greater than the threshold, a coordinate system is established to identify the frequency of occurrence of each parameter value in the parameter set at the target timestamp, a line graph model is established according to the frequency of occurrence of each parameter value in the parameter set at the target timestamp, the change characteristics of the line graph model are extracted, and the change characteristics of the line graphs of the parameter sets of each target timestamp are fused.

9. The big data-based power grid dynamic parameter identification system according to claim 8, characterized in that: The step S32 further comprises the following steps: Step S321: retrieve the real-time operation status data of the equipment, and compare the equipment operation status data predicted by the parameter identification theoretical model. When the real-time operation status of the equipment does not match the predicted operation status, identify the topology of the power grid system and the grid-connected data. If the current power system topology and grid-connected power have not changed, mark the current equipment as faulty. Otherwise, continue to detect. Step S322: retrieve the parameter prediction results of the parameter identification theoretical model and the results of the time-frequency analysis, calculate the difference between the parameter values ​​of the above prediction results, if the difference between the parameters of the prediction results is less than the threshold, select the average value of the prediction results as the judgment value, retrieve the real-time power grid parameters, calculate the difference with the judgment value, if the difference is less than the minimum threshold set by the system, mark the power system operation as normal, otherwise mark the power system operation as faulty, otherwise compare the actual parameters with the parameter prediction results of the parameter identification theoretical model and the results of the time-frequency analysis respectively; Step S323: obtaining the number of devices and grid-connected power in the real-time power system topology structure, and retrieving the corresponding influencing coefficients β and κ on the power grid parameters in the database according to the number of devices and grid-connected power in the power system topology structure; Step S324: When the difference between the parameter value of the parameter prediction result of the parameter identification theoretical model and the parameter value of the result of the time-frequency analysis is greater than the threshold, the result of the time-frequency analysis is obtained, and the actual parameter value corresponding to the analysis result is calculated by the formula Q=T·α·β·κ, where Q represents the actual parameter value corresponding to the time-frequency analysis result, and the real-time power grid parameter value is obtained. The difference F1 and F2 between the real-time power grid parameter value and the actual parameter value and the parameter value of the parameter prediction result of the parameter identification theoretical model are calculated respectively. If F1 and F2 are both less than the threshold, the current power system state is marked as normal. If F1 is less than the threshold and F2 is greater than the threshold, the time domain analysis model is marked as faulty. If F1 is greater than the threshold and F2 is less than the threshold, the parameter identification theoretical model is iteratively updated. Otherwise, the power system is marked as faulty. By comparing with the prediction results of the parameter identification theoretical model and the time domain analysis results respectively, the contingency can be further reduced, and a large deviation in the system identification results can be avoided, thereby further improving the accuracy of the system.

10. The big data-based power grid dynamic parameter identification system according to claim 9, characterized in that: In step S4, fog nodes around the data acquisition device are retrieved, the remaining bandwidth and data cache remaining capacity of the target fog node are identified, the corresponding upload speed V1 in the database is retrieved according to the bandwidth, the corresponding cache speed influence coefficient μ in the database is retrieved according to the data cache remaining capacity, the data collected by the data acquisition device is obtained, the number of bytes of the data is identified, and the amount of fog node cache occupied by the data to be transmitted is calculated by the formula Where H represents the amount of fog node cache occupied by the data to be transmitted, W represents the number of bytes of data to be transmitted, and V2 represents the cache speed of the target fog node. If the remaining capacity of the data cache of the target fog node is less than H, the fog node with the remaining capacity of the data cache that meets the conditions in the neighboring fog nodes is retrieved and marked as the first fog node. The fog node port code in the data port of the data acquisition device is cleared, and the port code in the first fog node is identified for connection. After the connection is completed, data transmission is performed.

Images